Warped-RE: Low-Cost Error Detection and Correction in GPUs

• Published in: 2015 45th Annual IEEE/IFIP International Conference on Dependable Systems and Networks pp. 331 - 342
• Main Authors: Abdel-Majeed, Mohammad, Dweik, Waleed, Hyeran Jeon, Annavaram, Murali
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2015
• Subjects: Benchmark testing; Error correction; error detection; GPUs; Graphics processing units; Instruction sets; Redundancy; Transient analysis; Tunneling magnetoresistance; value similarity

Graphics processing units (GPUs) are now the dominant computing fabric within many supercomputers. As such many mission critical applications run on GPUs, which demand stringent reliability and computational correctness guarantees from GPUs. Prior approaches to GPU reliability have tackled solely ei...